Science and Politics at the FDA

Pharmaceutical Executive: What do you think is the biggest misconception inside the pharma industry about the FDA?

Daniel Carpenter: There are two. The first is that people tend to see the FDA as a monolith, when in fact it's an organization with sharp internal disagreements about process and outcomes. The second myth is that the FDA is power hungry. FDA officials usually want authority when they think they can use it for public health purposes, but there are also a lot of searching discussions within the agency about what constitutes a credible line of action, given a particular regulatory problem.

Photo: Stu Rosner Photography

PE: The title of your book—"Reputation and Power"—highlights two themes at the heart of your approach to understanding the FDA. The reference to "reputation" may surprise many readers. As you explain in the book, reputation offers an alternative to the two most common (and opposing) ideological views—"regulatory capture" (the FDA as a servant to industry and its interests) and "regulatory overreach" (the FDA as an obstacle to industry and its innovations). Why this emphasis on reputation?

DC: A reputation is any set of beliefs that an audience holds about an individual or organization. What makes things complicated is that the audience is not homogeneous but a diverse population with varying interests and histories. So a reputation usually is a multifaceted set of beliefs. The same organization can have a good cop face, a bad cop face, or a facilitator face, and can show different faces to different audiences.

A reputation-based perspective succeeds where "capture" and "overreach" hypotheses fail. The "capture" explanation fails to account for most of the FDA's history, as many of the FDA's growing powers were strongly opposed by the pharmaceutical industry and the American Medical Association. An "overreach" hypothesis ignores the FDA's record of flexibility on a host of drugs, from HIV to cancer and even alternative medicine.

I'm not saying that a reputation-based account gets everything right, but it suggests that under some conditions the agency's approach to pharmaceutical innovation will be stringent and cautious; under other conditions, the FDA will be facilitative of drug development and will emphasize safety less (postmarket regulation is an example, until recently).

PE: What is the relationship between the political context of "power" and regulation as it plays out at the FDA?

DC: I understand power in three ways: directive, gate-keeping, and conceptual. Directive power is the ability of the FDA to order companies to do things such as inspections and manufacturing requirements. Gate-keeping power is the power that results from the FDA's ability to veto a new drug. When companies abandon what may appear to be an apparently promising and profitable compound in Phase II because they think that it will not meet the FDA's approval criteria, we rarely see that reported in the news. Yet the FDA's power has been exercised nonetheless, because companies and medical researchers are anticipating the FDA's behavior to inform their decisions.

Conceptual power is the ability of the FDA to shape the vocabulary, methods, and tools of medicine and drug development. The three-phase system of clinical trials and the architecture of bioequivalence are examples where the FDA officials established processes that have been replicated throughout the world, affecting the very nature of science.

PE: You characterize the pharma industry as historically intimidated by the FDA, for example adopting the agency's increasingly stringent preapproval rules for how drugs are tested. If that is true, why is there such a widespread perception that the FDA is in the pocket of pharma?

DC: Most of the book is about the era from the 1930s to the early 1990s, when intimidation held true. Yet it is fair to say that things have changed. Perceptions matter, and when the agency reversed its longstanding position on pre-emption, it made the agency look inconsistent and obsequious.

User fees have added to the perception problem; it's not just me saying that, but [former general counsel at the FDA] Peter Barton Hutt, too.

The solution is not clear. Marcia Angell and others would have us do away with the fees altogether. I tend to favor Rep. Maurice Hinchey's (D-NY) solution, which is to keep the revenue, be honest and call it a tax, and sever the contractual tie between the taxes and agency performance.

PE: You write that FDA officials view some pharma companies more favorably than others. For example, Merck was seen as having considerable scientific integrity, and this may have swayed the FDA officials to approve Vioxx despite early safety signals. What determines a pharma company's cachet at the FDA?

DC: Scientific integrity is often witnessed through full disclosure—the capacity to admit weaknesses in one's own arguments or data. Assuming that the problems identified by FDA drug reviewer Dr. Thomas Marciniak in the Avandia case [reporting that at least a dozen patients on the drug in a major clinical trial had suffered serious heart problems but were not counted by GSK as adverse events] could have been identified by GSK officials beforehand, the company would have fared far better if it had admitted to these problems and focused on showing alternative interpretations of its results.

When you present a study as fully robust, it's easy for someone to cherry-pick the problems. Every study, and every drug, will have its "soft underbelly," so why not be rigorous about disclosing it? That kind of circumspection could serve the pharmaceutical industry well in its dealings with the FDA, the public, and the various scientific and medical professions.

PE: The reputation of the FDA has been tarnished over the past decade. In your analysis, this decline in trust coincides with the ascendancy in national politics of an antiregulatory ideology. This culminated during George W. Bush's administration with the appointment of FDA officials to carry out that agenda. What effect did this have on the career staff at the FDA?

DC: It had several effects. First, a lot of very talented people left the agency, and that damage will take a long time to repair. Second, a number of the people who remained began to adapt in different ways to the new reality. Some waited it out, others changed their minds. A large number of people were demoralized. The evisceration of regulatory capacity under the Bush administration is an inescapably sad tale.

PE: President Obama's top FDA appointees, Margaret Hamburg and Josh Sharfstein, came to the agency with distinguished records as public health officials and as previous heads of big city health departments. After their first 18 months in office, both have remarked at how long it takes for even small changes to work their way through the system. Is this just typical bureaucratic pushback or is some other kind of resistance at work?

DC: What may seem like bureaucratic pushback or intransigence is often attributable to the rule of law. Many of the FDA officials I've met are nervous about their obligations and limits under the law. In a constitutional system like ours, it can be a good thing when change comes slowly to administrative and regulatory agencies. Ultimately, governance of science or industry has to be consistent with the public good as encoded in law and regulation. It's a point that often gets ignored in the debates.

PE: After watching Commissioner Hamburg at the helm for 18 months, what kind of a job do you think she is doing?

DC: I'm impressed with her. One thing the commissioner has to do is to make good appointments and to communicate credibly. I think most would agree that Hamburg does both. Unlike the commissioners under George W. Bush, she understands that the FDA's credibility has been tarnished, so she is not overstating the agency's case on scientific or policy matters.

In addition, Dr. Hamburg and Dr. Sharfstein are pursuing a much-needed transparency initiative. With transparency, there are risks. As Hamburg's unprecedented decision to post on the FDA's website the conflicting memos of various FDA officials about what to do in the Avandia decision shows, by not presenting a united front to the public and the scientific community, people will perhaps begin to view the FDA as characterized by massive infighting. That said, in an age when anyone can tweet, blog, or talk to a reporter, I think she understands that there is a greater risk from trying to present a united front when there isn't one.

PE. In the New York Review of Books, Marcia Angell allowed that your book is "exhaustively documented" and "ambitious," but she was critical of you for neglecting what she perceives as the central problem at the FDA: "How to ensure that CDER, free of industry influence, protects the public from unsafe, useless drugs." Why did you decide not to identify "problems" and recommend "solutions" after doing such a deep dive into the FDA?

DC: I'm not against making judgments. I agree with several of her suggestions, such as a fuller clinical trial database and faster generic approvals. But I believe in the value of an analysis that tries to understand the FDA for what it is, rather than what it should be.

PE: It is generally said that the FDA's priorities swing on a pendulum, between approving more drugs as fast as possible, and rejecting more drugs due to safety fears. Do you think that the "pendulum" metaphor is accurate?

DC: Pendulum metaphors should always be regarded with caution. The binary tradeoff between safety and access is far overplayed in our discussions about pharmaceutical policy. The transformative issues lie not in "safety versus innovation," but in providing good information that gives confidence to patients, physicians, health providers, and others who believe that the drug supply is characterized by safety and efficacy. My hypothesis is that the FDA's testing requirements establish not only a safer but also a more trustworthy marketplace.

PE: We seem to be entering an era where confidence in drug safety/efficacy evidence development and assessment is under assault due to a combination of factors, including the growing public accessibility of data, which leads to much expert "second guessing" of the decisions of advisory committees; the development of more sophisticated tools to analyze ever larger sets of data; and the effort by drug companies to be more transparent, which paradoxically has exposed corruption in their own disclosure processes. How can the FDA preserve its critical role as a gatekeeper if its authority to make even the most informed assessment of the data is increasingly undermined?

DC: You're right that the FDA and drug companies no longer have an oligopoly on possession of the data about drugs and those who test and market them. In terms of the FDA as gatekeeper, I suspect you will see more of what we saw in the Avandia decision, namely the FDA acknowledging that there are different hypotheses and arguments about the drug in question, and offering a rationale for its decisions. We live in a data-rich world where everyone is going to get second-guessed. But we still need a regulator to make final decisions according to public health criteria.

And as more second-guessing is done, I suspect you will see greater skepticism about the second-guesses themselves. The role of meta-analyses, epidemiologic studies, and even certain forms of randomized clinical trials is coming in for greater criticism. The FDA has issued its own guidance about meta-analyses, which suggests more doubt about their value. The agency is also trying to invest heavily in pharmacoepidemiology and statistics so that it can be a pivotal player in the public and scientific debates.

PE: How do you think the increasing globalization of the pharma market will impact the ability of the FDA to do its job?

DC: Globalization can weaken the FDA but also strengthen it. As other nations' drug markets grow larger, the pivotal nature of the FDA's gate-keeping power will decline, as it becomes less and less vital to get into the US market for drug companies to make a profit.

Another globalization effect is the growth of other nations' regulatory capacities; for safety's sake alone, we might welcome this. Americans' drug safety will increasingly depend upon the ability of regulators in India, China, and other countries to police the good manufacturing practices and quality of drugs produced there. The FDA knows this and is aggressively trying to help boost capacities and proselytize its models and practices overseas.

That said, the FDA model has been copied worldwide. As regulators in other countries begin to develop independent governmental capacities, they are seeking direct assistance from the FDA and more quietly copying (sometimes resisting) the FDA's decisions, structures, and processes.